This proposal will determine whether free or bound endogenous estrogens or androgens or their ratios are independently predictive of subsequent cardiovascular disease risk in a cohort of 2,330 white men aged 30-79 years who have been followed for 10 years. Relevant covariates determined by standardized protocol at the baseline evaluation (1972-74) include personal and family history of heart attack, heart failure, stroke, hypertension and diabetes; personal history of cigarette smoking and current use of antihypertensives, lipid lowering drugs or diet, and exogenous sex hormones; height, weight, blood pressure; and fasting plasma cholesterol, tryglyceride and glucose. Frozen plasma obtained at the baseline evaluation will be assayed in an endocrinology research laboratory for estrone, estradiol, androstenedione, testosterone, dehydroepiandrosterone and its sulfate, and sex hormone binding globulin. Followup data to the present (1985) includes 99.6% ascertainment for vital status (with a death certificate obtained for all decedents) and a self-report of morbidity. Fatal cardiovascular disease and self-reported morbidity are validated in a subset. This research will provide a unique prospective population based study of endogenous sex hormone levels as predictors of incident cardiovascular disease, ischemic heart disease, stroke and all cause mortality, after adjusting for the effect of age and other heart disease risk factors.